sign in sign up
<<
Home , Postural drainage

Netherlands Journal of Critical Care Submitted July 2015; Accepted December 2015

ORIGINAL ARTICLE

Intrapulmonary with autogenic drainage and ventilator-associated Gram-negative infection: A pilot study

H. Spapen, M. Borremans, M. Diltoer, J. De Regt, C. Bruggemans, P.M. Honoré Intensive Care Department, University Hospital, Vrije Universiteit Brussels, Brussels, Belgium

Correspondence H. Spapen – [email protected]

Keywords - chest physiotherapy, intrapulmonary percussive ventilation, assisted autogenic drainage, infection-related ventilator-associated complications, gram-negative

Abstract Background Background Intrapulmonary percussive ventilation with assisted autogenic CPT in mechanically ventilated patients primarily aims to drainage physiotherapy (IPV-AADP) is a compelling form facilitate clearance of airway secretions in an attempt to of (CPT) in mechanically ventilated prevent , tracheobronchial infection, and critically ill patients. We evaluated the effect of IPV-AADP on ventilator-associated (VAP).[1,2] For this purpose, the occurrence of Gram-negative infection-related ventilator- ICU physiotherapists have a large armamentarium of physical associated complications (IVACs). and mechanical means at their disposal. However, CPT-driven secretion management is not standardised among ICUs and Methods varies from simply providing appropriate humidification and Patients requiring for at least 48 as-needed airway suctioning to a more complex ‘multimodality’ hours were randomly assigned to receive either IPV-AADP, approach including percussion, manual or ventilator conventional CPT (CoCPT) or no CPT (NoCPT). Standard hyperinflation, chest wall vibrations, postural drainage, and rib- institutional prevention measures of ventilator-associated cage compression. infection were guaranteed in all subjects. The study endpoint Intrapulmonary percussive ventilation (IPV) delivers very was the presence of IVACs as documented according to the small bursts of within a frequency range of 60- Centers of Disease Control 2011 Working Group guidelines. 600 cycles/minute. By providing a convective front of gas to the Statistical analysis used non-parametric tests for independent distal airways, IPV provides a more homogenous distribution samples and Fisher’s exact test to compare treatment groups. of alveolar ventilation, promotes alveolar recruitment, helps to Results: Forty-five patients (24 males, 21 females) were enrolled ‘unstick’ in small and middle-sized airways, and propels with 15 subjects included in each study arm. IPV-AADP patients secretions cephalad to the central airways.[3] Encouraging results were younger (46±17 years) than CoCPT (62±18 years) and with IPV have been obtained in patients with .[4] NoCPT (64±16 years) subjects (p=0.014; IPV-AADP vs. CoCPT For more than 20 years, IPV physiotherapy has replaced all other and NoCPT) but APACHE II scores were comparable between forms of CPT in our ICU. With few exceptions, this technique is the groups (20±8, 23±10 and 21±6 respectively for IPV-AADP, routinely performed in all intubated and mechanically ventilated CoCPT and NoCPT subjects, p=NS). Gram-negative IVACs patients, regardless of their underlying clinical condition. were diagnosed in two patients (13%) in the IPV-AADP group, The efficacy of IPV could be enhanced by combining it with seven patients (47%) in the CoCPT group and seven patients autogenic drainage. Essentially, autogenic drainage is a (47%) in the NoCPT group (p=0.10; IPV-AADP vs. CoCPT and ‘concentration intensive’ respiratory self-drainage technique NoCPT). that explores and utilises the most optimal expiratory airflow to mobilise secretions.[5] It consists of three phases: loosening Conclusion peripheral secretions by at low lung volumes, In this small pilot study, adjunctive IPV-AADP tended to collecting secretions from central airways by breathing at low to decrease the occurrence of ventilator-associated Gram-negative mid lung volumes, and finally expelling secretions by breathing at infection as compared with CoCPT and NoCPT. mid to high lung volumes. Today, it is used to treat patients with

6 NETH J CRIT CARE - VOLUME 24 - NO 2 - MARCH 2016 Netherlands Journal of Critical Care Intrapulmonary percussion with autogenic drainage and ventilator-associated Gram-negative infection: A pilot study

large amounts of thick mucus. In assisted autogenic drainage, USA) using pulsatile percussions between 80 and 350 cycles the respiratory physiotherapist uses exhaled air to remove per minute. Gas was delivered at adjustable pressure through mucus from the airways of more severely ill patients who are a non-gated sliding venturi connected to the endotracheal not capable of performing autogenic drainage independently. tube. A pressure-controlled tidal volume breath was insufflated Combining IPV with assisted autogenic drainage physiotherapy at regular time intervals to ensure adequate CO2 elimination. (IPV-AADP) might represent a more powerful form of CPT FiO2 and PEEP levels were kept identical to those used on the than the currently used ‘multimodality’ techniques for adjuvant ventilator. For assisted autogenic drainage, gentle bimanual prevention of respiratory infection in mechanically ventilated compression was progressively exerted on the patient’s critically ill patients. We therefore assessed the effect of and maintained at a level where secretions could be ‘heard and providing either no CPT (NoCPT), conventional CPT (CoCPT) felt’. Subsequently, IPV-generated expiratory flow was used to or IPV-AADP on the occurrence of Gram-negative infection- expel the collected secretions. At the end of each IPV-AADP related ventilator-associated complications (IVACs) in this session, endotracheal suction was performed. population. In all three groups, identical standardised measures for prevention of ventilator-associated tracheobronchitis or Methods pneumonia were undertaken: ventilation with at least 5 cm H2O The study was approved by the Health Research Ethics PEEP, no routine change of ventilator circuits, humidification Committee of the University Hospital Brussels (B.U.N. using a heat-moisture device, semi-recumbent body 14320084389). Oral informed consent to initiate the study positioning at an angle of at least 25°, continuous maintenance was obtained from the patient’s next-of-kin. Mechanically of endotracheal tube cuff pressure at 30 cm H2O, periodic ventilated medical, surgical and/or trauma patients with verification of residual gastric volume, daily dental hygiene, negative tracheobronchial cultures were consecutively enrolled. oral cleaning with 1% chlorhexidine every eight hours, a strict Exclusion criteria for the study were: presence of lung infiltrates, sedation protocol aiming at minimal sedation according to a ventilation for less than 48 hours, community-acquired or plain dedicated sedation score and ensuring daily ‘wake-up’ calls. aspiration pneumonia, nosocomial or healthcare-acquired All patients received H2-blockers until adequate enteral infection, immunosuppressive disease, neutropenia, any form nutrition was achieved. Specific types of endotracheal tubes of immunosuppressive therapy including steroids, worse short- (e.g. subglottic aspiration, silver-coated) or cuff material (e.g. term prognosis, or a contraindication for IPV-AADP (i.e. polyurethane) were not used. Patients received no selective undrained pneumothorax, post sternotomy or , rib digestive decontamination treatment. fractures). During the study, all patients were ventilated in the Study endpoint was a documented Gram-negative IVAC. IVACs pressure-controlled mode (Evita XL, Dräger Medical). Inspired were defined according to the Centres of Disease Control 2011 oxygen fraction and positive end-expiratory pressure (PEEP) Working Group guidelines.[6] Briefly, on or after calendar Day 3 were adapted to obtain a PaO2 ≥60 mmHg. Inspiratory pressure of mechanical ventilation and within two calendar days before aimed at tidal volumes of 6-8 ml/kg and was limited at 30 cm or after the onset of worsening oxygenation, patients had to have H2O. Patients were randomly assigned according to a computer- either fever >38°C or a white blood cell count ≥12,000/mm³ or steered permuted block design to receive NoCPT, CoCPT or ≤4000/mm³ and a new antimicrobial agent started for at least IPV-AADP treatment. CoCPT consisted of expiratory chest four days. Additionally, the following criteria had to be met: wall percussion and vibration, positioning, rib springing, aerosol at least one purulent respiratory secretion specimen collected therapy, and airway suctioning. The NoCPT group underwent from the , bronchi, or containing ≥25 neutrophils passive mobilisation, aerosol therapy and tracheobronchial and ≤10 squamous epithelial cells per low power field (or aspiration. Aerosols contained 4 ml hypertonic saline 3% and corresponding semi-quantitative results) and a positive Gram- occasionally ipratroprium bromide, and were administered negative culture (qualitative, semi-quantitative or quantitative) three times daily by metered-dose inhalators connected to a of sputum, endotracheal aspirate or spacer. Closed suction was performed for no longer than 15 fluid. IVACs were considered early-onset when they were seconds using pressures ranging from 90-120 mmHg. IPV- diagnosed during the first four days of mechanical ventilation AADP and CoCPT were performed by two dedicated and and categorised as late-onset after four days. Statistical analysis skilled respiratory therapists. Sessions were delivered twice used non-parametric tests for independent samples and Fisher’s daily for 20 minutes, also during the weekend. During CPT, exact test to compare treatment groups. continuous intravenous analgosedation with remifentanil and propofol was provided in doses to obtain a -4 to -5 sedation level Results on the Richmond Agitation-Sedation Scale. IPV was delivered Forty-five patients (24 males, 21 females) were enrolled by switching the patient from the ventilator to a high-frequency with 15 subjects included in each study arm. Demographic pulse generator (Bird IPV 2 Percussionaire®, Sand Point, Idaho, characteristics, admission diagnosis and culture results for

NETH J CRIT CARE - VOLUME 24 - NO 2 - MARCH 2016 7 Netherlands Journal of Critical Care Intrapulmonary percussion with autogenic drainage and ventilator-associated Gram-negative infection: A pilot study

Table 1. Patient characteristics and microbiology. after reanimation. Before the start of the study, six CoCPT, Gender Age Admission diagnosis Microbiology six NoCPT and four IPV-AADP patients received antibiotic CoCPT treatment (p=NS between groups). IPV-AADP patients were F 82 Liver abscess younger (46±17 years) than CoCPT (62±18 years) and NoCPT F 83 Urosepsis (64±16 years) subjects (p=0.014; IPV-AADP vs. CoCPT and M 63 Haemorrhagic shock NoCPT) but APACHE II scores were comparable between the F 47 Necrotising fasciitis groups (20±8, 23±10 and 21± 6 respectively for IPV-AADP, M 48 Tetanus M 40 Cardiogenic shock P. vulgaris; S. marcescens CoCPT and NoCPT subjects, p=NS). Gram-negative IVACs M 80 Intracranial bleeding E. coli; P. mirabilis were diagnosed in two patients (13%) in the IPV-AADP group, M 54 Peritonitis seven patients (47%) in the CoCPT group and seven patients M 80 Intracranial bleeding E. coli; M. morganii (47%) in the NoCPT group (p=0.10; IPV-AADP vs. CoCPT M 67 Peritonitis and NoCPT). Time from start of the study till occurrence of a M 51 Intracranial bleeding E. coli Gram-negative IVAC ranged from 4-11 days. Early-onset IVAC M 76 Peritonitis was detected in four NoCPT patients, one CoCPT-treated F 69 Cardiac arrest E. coli subject and in no IPV-AADP patients (p=NS between groups). M 22 Head trauma K. pneumoniae Appropriate antibiotics were prescribed to treat all documented M 67 Stroke IVACs. Survival was not different between the groups and NoCPT unrelated to the identified Gram-negative microorganism. F 70 Intracranial bleeding K. pneumoniae F 80 Epiglottitis P. aeruginosa Clinical or device-related adverse events were not reported with M 35 Polytrauma E. coli any form of CPT. F 76 Sepsis M 73 Haemorrhagic shock Discussion M 60 Osteomyelitis Through the years, CPT has become an integral part of the M 74 Sepsis P. aeruginosa multidisciplinary approach to critically ill ventilated patients. F 49 Intracranial bleeding E. coli; K. pneumoniae However, there is a dearth of high-level evidence to support any F 77 Sepsis E. coli particular CPT technique. F 50 Polytrauma Studies that assessed the usefulness of CPT in ICU patients M 82 Head trauma mainly evaluated mortality rate or clinical outcomes such as F 73 Polytrauma incidence of VAP, ventilator-free days, or length of ICU and M 57 Meningitis M 73 Encephalitis hospital stay. All relevant prospective and controlled studies F 33 Intracranial bleeding H. influenzae are summarised in table 2. Ntoumenopoulos et al. studied IPV-AADP 46 mechanically ventilated trauma patients (22 CPT and 24 F 65 Intracranial bleeding controls). CPT consisted of twice-daily lung hyperinflation M 61 Intracranial bleeding and postural drainage. Patients in both groups received routine M 38 Cardiac arrest nursing care, including at least two-hourly patient turning and M 21 Polytrauma airway suctioning. CPT was not associated with a reduced M 63 Cardiogenic shock E. cloacae incidence of VAP according to clinical and radiological criteria. M 36 Encephalitis [7] The same investigators presented another study in 60 critically F 29 Intracranial bleeding ill mechanically ventilated medicosurgical patients (24 CPT vs. F 71 Intracranial bleeding 36 controls). Here, CPT comprised twice daily positioning, M 53 Cardiac arrest expiratory chest wall vibrations and airway suctioning. VAP F 60 Hepatic encephalopathy F 69 Peritonitis occurred in 14 (39%) control patients and in two (8%) CPT [8] F 34 Status epilepticus subjects (OR=0.4; 95% CI 0.03-0.56; p=0.02). A difference in F 26 Cardiac arrest duration of ventilation, length of ICU stay, or mortality rate F 39 Intracranial bleeding was not observed in either of these studies.[6,7] Templeton and F 31 Encephalitis E. cloacae Palazzo evaluated 172 patients (87 CPT vs. 85 controls). CPT comprised twice-daily patient positioning, manual pulmonary CoCPT = conventional chest physiotherapy; NoCPT = no chest physiotherapy; IPV-AADP = intrapulmonary percussive ventilation with hyperinflation with vibration, rib springing, and airway assisted autogenic drainage physiotherapy; M = male; F = female. suctioning. VAP was observed in more patients receiving CPT than in control patients (35 vs. 25; p=0.13). Patients all treatment groups are depicted in table 1. The majority of receiving CPT had significant prolongation of median time to patients (49%) had primary cerebral pathology or were admitted become ventilator-free.[9] Pattanshetty and Gaude studied 101

8 NETH J CRIT CARE - VOLUME 24 - NO 2 - MARCH 2016 Netherlands Journal of Critical Care Intrapulmonary percussion with autogenic drainage and ventilator-associated Gram-negative infection: A pilot study

Table 2. Prospective controlled studies of chest physiotherapy in mechanically ventilated patients. Author (reference) Number of patients RPT intervention Result Ntoumenopoulos[7] 22 CPT MLH, postural drainage; bid No difference in VAP incidence and 24 Controls duration of ventilation between groups Ntoumenopoulos[8] 24 CPT Body positioning, expiratory chest wall Less VAP in CPT group 36 Controls vibrations, suction; bid Templeton & Palazzo[9] 87 CPT Body positioning, MLH, rib springing, Tendency for more VAP and prolonged 85 Controls chest wall vibrations, suction; bid ventilation in CPT group Pattanshetty & Gaude[10] 50 CPT Body positioning, chest wall vibrations, Less VAP, more successful weaning 51 Controls suction; bid and lower mortality in CPT group; no difference in duration of ventilation or length of ICU stay Pattanshetty & Gaude[11] 87 CPT Body positioning, MLH, chest wall No difference in VAP incidence 86 Controls vibrations, suctioning; bid (controls between groups; prolonged duration of received MLH and suction hospitalisation in CPT group Patman[12] 72 CPT Body positioning, MLH, suction; x6/day No significant difference between 72 Controls groups for any outcome

CPT = chest physiotherapy; MLH = manual lung hyperinflation; VAP = ventilator-associated pneumonia; bid= twice daily; ICU = intensive care unit adult mechanically ventilated patients. Twice-daily manual methodological quality and range of samples in these hyperinflation and suctioning were administered to 51 control studies preclude statistical analysis of pooled results. Also, patients. Fifty CPT patients additionally received positioning the ‘multimodality’ CPT applied in most studies makes and chest wall vibrations. As compared with controls, CPT it impossible to determine the effectiveness of individual was associated with a highly significant decrease in VAP rate treatment components. Evaluation of a potential impact of and reduced mortality (24% vs. 49%; p=0.007).[10] However, a CPT on outcome variables such as ICU/hospital length of stay subsequently published study by the same authors including and mortality is most likely flawed by differences in underlying 173 patients found no difference in VAP incidence and a disease type and severity and concomitant comorbid disorders. longer duration of hospitalisation in the 87 subjects allocated In all but one[12] study, diagnosis of VAP relied on clinical, to a similar multimodality CPT.[11] Patman et al. studied 144 biological, and radiographic criteria, sometimes bundled in mechanically ventilated patients with brain injury (72 CPT and a score. However, pooled specificity and sensitivity of such 72 controls). Diagnosis of VAP required the presence of positive assessment is low[14] and inter-observer variability in the quantitative cultures obtained by non-bronchoscopic lavage. diagnosis of VAP is extremely high.[15] Our study is the first to use CPT consisted of six treatments (i.e. positioning, manual lung a novel approach of estimating ventilator-associated infection hyperinflation, and airway suctioning) in each 24-hour period. based on documenting purulent secretions and microbiological No difference was observed in VAP incidence between the findings but eliminating both the low specificity of chest X-ray groups (14 CPT vs. 19 controls; p=0.32). Although CPT tended interpretation and the variability in ordering and collecting to reduce the duration of mechanical ventilation and length of lower specimens. Although this approach of ICU and hospital stay, none of these outcomes reached was essentially designed for surveillance purposes, it is more in statistical significance.[12] Finally, Castro et al. compared CPT keeping with traditional clinical constructs of VAP. consisting of at least four treatment sessions over a 24-hour Our study has important shortcomings and limitations. First, period with CPT consisting of only one visit over a six-hour as a pilot project, it is heavily underpowered and thus cannot period in 146 patients (73 subjects in each group). CPT was determine whether adding CPT with IPV-AADP to a standard similar for both periods and consisted of body positioning, VAP prevention program may cause fewer Gram-negative manual thorax percussion, and suction. Patients undergoing IVACs. Second, bias introduced by the heterogeneity in intensive CPT had significantly less duration of ventilation admission diagnoses, the incompletely documented medico- and ICU stay, developed fewer respiratory infections, and had surgical patient histories, the concomitant antibiotic therapy lower mortality.[13] However, this study compared CPT in two for non-pulmonary infectious disease, the inter-individual different hospitals. Important baseline patient characteristics difference in duration of stress ulcer prophylaxis and enteral such as degree of organ failure, sedation level, and coma scale nutrition policy, and the younger age of the IPV-AADP patients were substantially or significantly different between patients may have skewed the study results. Third, the high prevalence which makes any valuable comparison doubtful. of acute cerebral pathology renders evaluation of outcome Overall, the evidence emanating from these trials on the parameters such as duration of mechanical ventilation or ICU usefulness of CPT in intubated and mechanically ventilated length of stay irrelevant. Fourth, it can be argued that reporting patients is conflicting. Differences in CPT interventions, Gram-negative IVACs is an inappropriate study endpoint.

NETH J CRIT CARE - VOLUME 24 - NO 2 - MARCH 2016 9 Netherlands Journal of Critical Care Intrapulmonary percussion with autogenic drainage and ventilator-associated Gram-negative infection: A pilot study

Despite being proposed as an algorithm to streamline diagnosis, References the Centres of Disease Control definition of ventilator-associated 1. Volsko TA. : Finding the evidence. Respir Care. conditions is found to insufficiently capture VAP.[16] Conversely, 2013;58:1669-78. 2. Stiller K. Physiotherapy in intensive care. An updated systematic review. Chest. measures to decrease the incidence of VAP do not necessarily 2013;144:825-47. affect IVAC rates.[17] However, our study results should be 3. Kallet RH. Adjunct therapies during mechanical ventilation: Airway clearance techniques, therapeutic aerosols, and gases. Respir Care. 2013;58:1053-71. interpreted within the pathogenic continuum of progressive 4. Varekojis SM, Douce FH, Flucke RL, et al. A comparison of the therapeutic Gram-negative tracheobronchial colonisation with bacterial effectiveness of and preference for postural drainage and percussion, intrapulmonary percussive ventilation, and high-frequency chest wall growth and increasing inflammation leading to ventilator- compression in hospitalized cystic fibrosis patients. Respir Care. 2003;48:24-8. associated tracheobronchitis and eventually VAP.[18] Ventilator- 5. Agostini P, Knowles N. Autogenic drainage: the technique, physiological basis and evidence Physiotherapy. 2007;93:157-63. associated tracheobronchitis appears to be an important risk 6. Magill SS, Klompas M, Balk R, et al. Developing a new, national approach to factor for VAP. Moreover, patients in the ‘ventilator-associated surveillance for ventilator-associated events. Crit Care Med. 2013;41:2467-75. 7. Ntoumenopoulos G, Gild A, Cooper DJ. The effect of manual lung hyperinflation tracheobronchitis phase’ do benefit from targeted antimicrobial and postural drainage on pulmonary complications in mechanically ventilated treatment in terms of more ventilator-free days and less trauma patients. Anaesth Intensive Care. 1998;26:492-6. 8. Ntoumenopoulos G, Presneill JJ, McElholum M, Cade JF. Chest physiotherapy [19] subsequent VAP Finally, the relative complexity, availability, for the prevention of ventilator-associated pneumonia. Intensive Care Med. and cost of IPV might be questioned. Yet, physiotherapists are 2002;28:850-6. 9. Templeton M, Palazzo MGA. Chest physiotherapy prolongs duration of rapidly trained to work with the device (personal experience) ventilation in the critically ill ventilated for more than 48 hours. Intensive Care and any ICU involved in high-frequency ventilation can easily Med. 2007;33:1938-45. 10. Pattanshetty RB, Gaude GS. Effect of multimodality chest physiotherapy in and safely implement IPV physiotherapy. prevention of ventilator-associated pneumonia: A randomized clinical trial. Indian J Crit Care Med. 2010;14:70-6. 11. Pattanshetty RB, Gaude GS. Effect of multimodality chest physiotherapy on the Conclusions rate of recovery and prevention of complications in patients with mechanical ventilation: A prospective study in medical and surgical intensive care units. CPT combining IPV with assisted autogenic drainage tends to Indian J Med Sci. 2011;65:175-85. reduce the occurrence of Gram-negative IVACs. The limited 12. Patman S, Jenkins S, Stiller K. Physiotherapy does not prevent, or hasten recovery from, ventilator-associated pneumonia in patients with acquired brain injury. number, heterogeneity, and baseline disease characteristics of Intensive Care Med. 2009;35:258-65. the studied patients preclude determining whether IPV-AADP 13. Castro AAM, Calil SR, Freitas SA, Oliveira AB, Porto EF. Chest physiotherapy effectiveness to reduce hospitalization and mechanical ventilation length of stay, hastens recovery, shortens duration of ventilation, or improves pulmonary infection rate and mortality in ICU patients. Respir Med. 2013;107:68- outcome. 74. 14. Shan J, Chen H-L, Zhu J-H. Diagnostic accuracy of clinical pulmonary infection score for ventilator-associated pneumonia: a meta-analysis. Respir Care. Disclosures 2011;56:1087-94. 15. Klompas M. Interobserver variability in ventilator-associated pneumonia All authors declare no conflict of interest. No funding or surveillance. Am J Infect Control. 2010:38:237-9. financial support was received. 16. Boyer AF, Schoenberg N, Babcock H, McCullen KM, Micek ST, Kollef MH. A prospective evaluation of ventilator-associated conditions and infection-related ventilator-associated conditions. Chest. 2015;147:68-81. 17. Muscedere J, Sinuff T, Heyland DK, et al. The clinical impact and preventability of ventilator-associated conditions in critically ill patients who are mechanically ventilated. Chest. 2013;144:1453-60. 18. Craven DE, Chroneou A, Zias N, Hjalmarson KI. Ventilator-associated tracheobronchitis. The impact of targeted antibiotic therapy on patient outcomes. Chest. 2009;135:521-8. 19. Agrafiotis M, Siempos II, Falagas ME. Frequency, prevention, outcome and treatment of ventilator-associated tracheobronchitis: systematic review and meta-analysis. Respir Med. 2010;104:325-36.

10 NETH J CRIT CARE - VOLUME 24 - NO 2 - MARCH 2016